CN103560974B - Method and device for maintaining tokens - Google Patents
Method and device for maintaining tokens Download PDFInfo
- Publication number
- CN103560974B CN103560974B CN201310514262.5A CN201310514262A CN103560974B CN 103560974 B CN103560974 B CN 103560974B CN 201310514262 A CN201310514262 A CN 201310514262A CN 103560974 B CN103560974 B CN 103560974B
- Authority
- CN
- China
- Prior art keywords
- token
- pir
- speed
- rate
- max
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/215—Flow control; Congestion control using token-bucket
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0894—Packet rate
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2408—Traffic characterised by specific attributes, e.g. priority or QoS for supporting different services, e.g. a differentiated services [DiffServ] type of service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
- H04L47/2458—Modification of priorities while in transit
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/31—Flow control; Congestion control by tagging of packets, e.g. using discard eligibility [DE] bits
Abstract
The embodiment of the invention discloses a method and device for maintaining tokens to solve the problem that burst of transmitted data is overlarge and to improve the QoS, and relates to the technical field of communication. The method includes the steps of obtaining the PIR of a network node, splitting the PIR into a first rate and a second rate, obtaining the number of the added first tokens corresponding to a first period according to the first rate, obtaining the number of the added second tokens corresponding to a second period according to the second rate, adding the tokens according to the first period and the number of the adding first tokens, and adding the tokens according to the second period and the number of the adding second tokens. The method and the device are suitable for the token adding scene in the token maintaining process.
Description
Technical field
The present invention relates to communication technical field, the method and apparatus that more particularly, to a kind of token is safeguarded.
Background technology
Differentiated service is ip qos(ip quality of service most widely used at present, ip service quality)
Model, using differentiated service network node send data during, need to regulate and control data by using token
Transmission rate.And it is provided with token bucket within network nodes, token is stored in token bucket, and network node sends certain every time
The data volume of size is accomplished by consuming a certain amount of token.But the capacity of token bucket is pre-set, therefore in token
After consumption, need periodically to add in token bucket token.For example, the memory capacity shared by token that each cycle is added
For:
pir*t=token
Pir(peak information rate, peak information rate) for network node message transmission rate；T is order
The interpolation cycle of board；Token by every time memory capacity shared by interpolation token.And the process that network node is safeguarded in token
The middle memory capacity shared by token needing record to be added in token bucket in a cycle, so that network node determines order
Whether all tokens being stored at present in board bucket can support the transmission of data.
In order to realize preferable qos, the transmission rate of control data, often select the little cycle to add token as far as possible,
As chosen the minimum period that hardware performance can be supported by, but select the little cycle to easily lead to token value and fractional part occur,
And the fractional part storing token value needs to take more storage resources.Prior art, in order to avoid the appearance of decimal, is chosen
The larger cycle adds token so that when pir is minimum, the token value of a cycle is also integer.
But, prior art chooses the larger cycle, when pir is very big, the token of a cycle interpolation
Token is excessive, easily causing and needs the data burst transmitting excessive, thus peak flow, therefore reducing qos.
Content of the invention
Embodiments of the invention provide the method and apparatus that a kind of token is safeguarded, can solve the problem that the excessive problem of burst, and
Improve qos.
For reaching above-mentioned purpose, embodiments of the invention adopt the following technical scheme that
In a first aspect, the embodiment of the present invention provides a kind of method that token is safeguarded, comprising:
Obtain the peak information rate (pir) of network node；
Described pir is split into first rate and the second speed, wherein, described first rate and described second speed total
With equal to described pir；
First token corresponding with the period 1 is obtained according to described first rate and adds quantity, according to the described second speed
Rate obtains second token corresponding with second round and adds quantity；
Add quantity according to described period 1 and described first token and add token, according to described second round and described
Second token adds quantity and adds token.
In conjunction with a first aspect, in the first possible implementation, methods described also includes:
Obtain the minimum refresh cycle that the hardware performance of described network node can be supported；
One setting value being more than or equal to the described minimum refresh cycle it is the described period 1.
Optionally, described described pir is split into first rate and the second speed, comprising:
Obtain l and k, and obtain n, n=l/(k*t1), wherein, l represents minimum message length, and k represents the precision of pir, t1Table
Show the described period 1；
Obtain w, wherein n=2 using described nw, or 2wIt is a smallest positive integral being more than n；
Determine 0 to the m bit of described pir, and obtain the binary number of w position to m position；
Binary number according to w position to m position obtains described first rate.
Further, described described pir is split into first rate and the second speed, also include:
In 0 to the m bit of described pir, obtain the binary number of w ' position to w-1 position, wherein, 0≤w '
＜ w；
Binary number according to w ' position to w-1 position obtains described second speed.
In conjunction with the first mode in the cards of first aspect and first aspect, in the possible implementation of second
In, methods described also includes:
Obtain tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent described pir corresponding cycle, tmax≥l/k；
Obtain token difference, described token difference=tokenmax- v1*t1*[tmax/t1], wherein v1Represent described first speed
Rate；
Obtain described second round using described token difference and described second speed.
Second aspect, the embodiment of the present invention provides the device that a kind of token is safeguarded, comprising:
First acquisition module, for obtaining the peak information rate (pir) of network node；
Split module, for described pir is split into first rate and the second speed, wherein, described first rate and institute
The summation stating the second speed is equal to described pir；
Second acquisition module, adds number for obtaining the first token corresponding with the period 1 according to described first rate
Amount, obtains second token corresponding with second round according to described second speed and adds quantity；
Add module, adds token for adding quantity according to described period 1 and described first token, according to described
Second round and described second token add quantity and add token.
In conjunction with second aspect, in the first possible implementation, described device also includes:
3rd acquisition module, the minimum refresh cycle that the hardware performance for obtaining described network node can be supported；
Period 1 acquisition module, the setting value for being more than or equal to the described minimum refresh cycle by is described the
One cycle.
Optionally, described fractionation module, comprising:
First acquisition unit, for obtaining l and k, and obtains n, n=l/(k*t1), wherein, l represents minimum message length, k
Represent the precision of pir, t1Represent the described period 1；
Second acquisition unit, for obtaining w, wherein n=2 using described nw, or 2wIt is a smallest positive integral being more than n；
3rd acquiring unit, for determining 0 to the m bit of described pir, and obtains w position and enters to the two of m position
Number processed；
First rate acquiring unit, obtains described first rate for the binary number according to w position to m position.
Further, described fractionation module also includes:
4th acquiring unit, in 0 to the m bit of described pir, obtains w ' position to the two of w-1 position
System number, wherein, 0≤w ' ＜ w；
Second speed acquiring unit, obtains described second speed for the binary number according to w ' position to w-1 position.
In conjunction with the first possible implementation of second aspect and second aspect, in the possible implementation of second
In, described device also includes:
4th acquisition module, for obtaining tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent that described pir is corresponding
Cycle, tmax≥l/k；
5th acquisition module, for obtaining token difference, described token difference=tokenmax- v1*t1*[tmax/t1], its
Middle v1Represent described first rate；
Second round acquisition module, for obtaining described second round using described token difference and described second speed.
The method and apparatus that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information of network node
Speed (pir), and pir is split into first rate and the second speed, obtain corresponding with the period 1 further according to first rate
First token add quantity, according to second speed obtain corresponding with second round second token interpolation quantity, according to first
Cycle and the first token add quantity and add token, add quantity according to second round and the second token and add token.With selection
The prior art adding token compared with large period is compared, the scheme that the present invention provides, and pir is divided for multiple different speed,
And add token respectively so that being less than according to the token that the speed single that each marks off adds according to each speed marking off
The token adding according to pir single, it is to avoid the data burst being transmitted by needs that the token that single adds excessively is led to
Excessive, thus reducing peak flow, therefore improve qos.
Brief description
For the technical scheme being illustrated more clearly that in the embodiment of the present invention, below by use required in embodiment
Accompanying drawing be briefly described it should be apparent that, drawings in the following description are only some embodiments of the present invention, for ability
For the those of ordinary skill of domain, on the premise of not paying creative work, can also be obtained other attached according to these accompanying drawings
Figure.
Fig. 1 is the example schematic adding token provided in an embodiment of the present invention；
Fig. 2 is the flow chart of the method that a kind of token provided in an embodiment of the present invention is safeguarded；
Fig. 3 is the flow chart of the method that another kind token provided in an embodiment of the present invention is safeguarded；
Fig. 4 is the flow chart of the method that another token provided in an embodiment of the present invention is safeguarded；
Fig. 5 is the structural representation of the device that a kind of token provided in an embodiment of the present invention is safeguarded；
Fig. 6 is the structural representation of the device that another kind token provided in an embodiment of the present invention is safeguarded；
Fig. 7 is that a kind of structure of specific implementation of device that another kind token provided in an embodiment of the present invention is safeguarded is shown
It is intended to；
Fig. 8 is the structure of another kind of specific implementation of device that another kind token provided in an embodiment of the present invention is safeguarded
Schematic diagram；
Fig. 9 is the structure of another specific implementation of device that another kind token provided in an embodiment of the present invention is safeguarded
Schematic diagram；
Figure 10 is the structural representation of the device that another token provided in an embodiment of the present invention is safeguarded.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is all other that those of ordinary skill in the art are obtained under the premise of not making creative work
Embodiment, broadly falls into the scope of protection of the invention.
Technical scheme provided in an embodiment of the present invention can be when limiting network node transmission data volume using token bucket, should
For adding in the scene of token.For example: a computer is as one of LAN network node, it is possible to use token
To limit the transmission rate of various data in LAN.This programme obtains the peak information rate (pir) of network node, and by institute
State pir and split into first rate and the second speed it is also possible to pir is divided into more speed, such as third speed, fourth rate
Deng further according to the token interpolation quantity in each speed acquisition cycle corresponding with each speed, according to each cycle and each week
Phase corresponding token adds quantity and adds token.For example: as shown in figure 1, in a large period, according to period 1 t1Add
First token add quantity be token1, and token1=v1*t1, v1For first rate, according to second round t2Second adding
It is token that token adds quantity2, and token2=v2*t2, v2For the second speed.
Embodiments provide a kind of method that token is safeguarded, as shown in Figure 2, comprising:
101, obtain the peak information rate (pir) of network node.
Wherein, pir can be expressed as the precision of n*pir, and n is positive integer, and the precision of pir can obtain network node
Learn during pir.
102, described pir is split into first rate and the second speed.
Wherein, the summation of first rate and the second speed is equal to pir.In actual applications, pir can be expressed as n*pir
Precision, n can use binary number representation, according to the n with binary number representation, by calculating, pir can be split into the
One speed and the second speed, first rate can use v1Represent, the second speed can use v2Represent, pir can also be split into more
Many speed, such as: pir is split into first rate, the second speed, third speed and fourth rate, and first rate, second
The summation of speed, third speed and fourth rate is equal to pir.
103, corresponding with the period 1 the first token interpolation quantity is obtained according to described first rate, according to described the
Two speed obtain second token corresponding with second round and add quantity.
Wherein, the first token interpolation quantity can use token1To represent, token1=v1*t1, v1For first rate, t1For
Period 1, the second token adds quantity and can use token2To represent, token2=v2*t2, v1For first rate, t2For second
Cycle.
104, add quantity according to described period 1 and described first token and add token, according to described second round and
Described second token adds quantity and adds token.
Wherein, the period 1 is corresponding with the first token interpolation quantity, and second round is corresponding with the second token interpolation quantity, the
One token adds quantity, the second token adds quantity and is integer, so it is not in that fractional part takies that token adds quantity
Storage resource.
The method that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir), and by pir split into first rate and the second speed, obtain corresponding with the period 1 the further according to first rate
One token adds quantity, obtains second token corresponding with second round according to the second speed and adds quantity, according to the period 1
Add quantity with the first token and add token, add quantity according to second round and the second token and add token.With from larger
The prior art that cycle adds token is compared, the scheme that the present invention provides, and pir is divided for multiple different speed, and presses
Add token respectively according to each speed marking off so that token that the speed single that marks off according to each adds be less than according to
The token that pir single adds, it is to avoid because the data burst that needs that the token that single adds excessively is led to transmit is excessive,
Thus reducing peak flow, therefore improve qos.
Optionally, on the basis of the method shown in Fig. 2, the embodiment of the present invention additionally provides the side that a kind of token is safeguarded
Method, specifically can be obtained first week according to the minimum refresh cycle that the hardware performance of network node can be supported by 202-203
Phase, as shown in figure 3, the method includes:
201, obtain the peak information rate (pir) of network node.
Wherein, 201 specific embodiments are identical with 101, will not be described here.
202, obtain the minimum refresh cycle that the hardware performance of described network node can be supported.
Wherein, the minimum refresh cycle that the hardware performance of network node can be supported is exactly the hardware performance in network node
Under conditions of can supporting, the minimum period for adding token can be chosen.
203, a setting value being more than or equal to the described minimum refresh cycle is the described period 1.
204, described pir is split into first rate and the second speed.
Wherein, 204 specific embodiments are identical with 102, will not be described here.
205, corresponding with the period 1 the first token interpolation quantity is obtained according to described first rate, according to described the
Two speed obtain second token corresponding with second round and add quantity.
Wherein, 205 specific embodiments are identical with 103, will not be described here.
206, add quantity according to described period 1 and described first token and add token, according to described second round and
Described second token adds quantity and adds token.
Wherein, 206 specific embodiments are identical with 104, will not be described here.
The method that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir), and by pir split into first rate and the second speed, can also be supported according to the hardware performance of network node
The minimum refresh cycle chooses the period 1, obtains first token corresponding with the period 1 further according to first rate and adds quantity,
Second token corresponding with second round is obtained according to the second speed and adds quantity, add number according to period 1 and the first token
Amount adds token, adds quantity according to second round and the second token and adds token.Add showing of token with from compared with large period
Technology is had to compare, the scheme that the present invention provides, pir is divided for multiple different speed, and according to each speed marking off
Rate adds token respectively so that being less than, according to the token that the speed single that each marks off adds, the order added according to pir single
Board, it is to avoid because the data burst that needs that the token that single adds excessively is led to transmit is excessive, thus reduce flow
Peak value, therefore improves qos.
Further, on the basis of method shown in Fig. 3, the embodiment of the present invention provides the side that a kind of token is safeguarded again
Method, 204 can be implemented as 304-309, and pir is split as first rate and the second speed；And by 310-312 determine with
Second speed corresponding second round, as shown in figure 4, the method includes:
301, obtain the peak information rate (pir) of network node.
302, obtain the minimum refresh cycle that the hardware performance of described network node can be supported.
303, a setting value being more than or equal to the described minimum refresh cycle is the described period 1.
304, obtain l and k, and obtain n, n=l/(k*t1).
Wherein, l represents described minimum message length, and k represents the precision of described pir, t1Represent the described period 1.Than
As: minimum message length is 8bit for 1byte, and precision k of pir is 16kbps, period 1 t1For 15.625 μ s, then n=
8bit/(16kbps*15.625μs)=32.Also such as: minimum message length is 8bit for 1byte, precision k of pir is
16kbps, period 1 t1For 15.625 μ s, then n=8bit/ (16kbps*16 μ s)=31.25.
305, obtain w using described n.
Wherein, n=2w, or 2wIt is a smallest positive integral being more than n.Such as: the n=32 obtaining in 304 example, then
2w=32, then w=5.Also such as: the n=31.25 obtaining in 304 example, then 2wIt is a smallest positive integral more than 31.25, then
2w=32, w=5.
306, determine 0 to the m bit of described pir, and obtain the binary number of w position to m position.
Wherein, in actual applications, pir can be expressed as the precision of n*pir, n be positive integer, n can be expressed as from 0 to
The binary number of m position, 0 is lowest order, and m position is highest order.Such as: pir=108223*16kpbs, then n is 108223, n table
The lowest order 0 being shown as the binary number that binary number is 11010011010111111, n is 1, and highest order 16 is 1, sets w
=5, then obtain the binary number of the 5th to the 16th.
307, the binary number according to w position to m position obtains described first rate.
Wherein, retain w position to the m position of the binary number of n of pir, m 1, to 0 position zero, obtains n1, first
Speed v1=n1* the precision of pir.Such as: obtain the 5th to the 16th of binary number of the n of pir in 306 example, with
When by the 4th to 0 position zero, the n obtaining1Be 11010011010100000 with binary number representation, i.e. n1=108192, then
To first rate v1=108192*16kpbs.
308, in 0 to the m bit of described pir, obtain the binary number of w ' position to w-1 position.
Wherein, 0≤w ' ＜ w.By the w position of the binary number of the n of pir to m position zero, retain the binary system of the n of pir
W ' position to the w-1 position of number, w ' 1, to 0 position zero, obtains n2If w ' 1 ＜ 0 is then it represents that pir is only divided into two
Speed.For example: in 306 example, 0 to 4 bit of pir is 11111, setting w '=0, w ' 1 ＜ 0, then pir extremely draws
It is divided into two speed, the n obtaining2It is 11111 with binary number representation.
309, the binary number according to w ' position to w-1 position obtains described second speed.
Wherein, according to 308, the second speed v2=n2*pir.Such as: in 308 example, only pir is divided into 2 speed
Rate, the n obtaining2Be 11111 with binary number representation, i.e. n2=31, then the second speed v2=31*16kpbs.
Wherein, pir not only can be divided into 2 speed it is also possible to divide more speed, such as third speed, the 4th
Speed etc..By that analogy, the computational methods of the speed such as third speed, fourth rate are similar with the algorithm of the second speed.Such as:
In 308 example, w '=1 can also be set, then the 1st to the 4th of the binary number of n of reservation pir, 0 position zero, obtain
N2Be 11110 with binary number representation, i.e. n2=30, then the second speed v2=30*16kpbs；Retain the binary system of the n of pir again
0 of number, the n obtaining3Be 1 with binary number representation, i.e. n3=1, then third speed v3=1*16kpbs.
310, obtain tokenmax, tokenmax=pir*tmax.
Wherein, tokenmaxRepresent in tmaxIn token add total quantity, tmaxRepresent described pir corresponding cycle, tmax
>=l/k, because pir can be expressed as the precision of n*pir, n is positive integer, thereby may be ensured that in cycle tmaxMiddle added
tokenmaxIt is the integral multiple of minimum message length, so in cycle tmaxIn, the token that addedmaxIt is not in fractional part
Point, also will not additionally take storage resource.Such as: pir is 108223*16kpbs, and minimum message length l is 8bit for 1byte,
Precision k of pir is 16kbps, tmax>=8bit/16kbps=500 μ s, sets tmax=500 μ s, then in cycle tmaxMiddle added
tokenmax=pir*tmax=108223*16kpbs*500μs=108223byte.Also such as: pir is 108223*16kpbs, minimum
Message length l is 8bit for 1byte, and precision k of pir is 16kbps, tmax>=8bit/16kbps=500 μ s, sets tmax=1000
μ s, then in cycle tmaxMiddle added tokenmax=pir*tmax=108223*16kpbs*1000μs=216446byte.
311, obtain token difference, described token difference=tokenmax- v1*t1*[tmax/t1].
Wherein, v1Represent described first rate, [tmax/t1] represent take tmax/t1Calculated integer part, that is,
In tmaxIn can be with t1Add the number of times of token for the cycle.Such as: pir=108223*16kpbs, minimum message length l are
1byte is 8bit, tmax=500 μ s, then in cycle tmaxMiddle added tokenmax=pir*tmax=108223*16kpbs*500μ
S=108223byte, period 1 t1For 15.625 μ s, calculated according to 304-307, obtained first rate v1=108192*
16kpbs, then can obtain token difference=tokenmax- v1*t1*[tmax/t1]=31byte.
312, obtain described second round using described token difference and described second speed.
Wherein it is considered to token difference is in tmaxIn can be with t2Add the number of times of token for the cycle, such as: in 311 example
In, set and pir is divided into 2 speed, then the second speed v2=31*16kpbs, because token difference is 31byte, Ke Yi
tmaxIn 1 time interpolation, then t2=31byte/1/(31*16kpbs)=500 μ s, 1 represents in tmaxIn with t2Add token 1 for the cycle
Secondary.Pir is divided into more speed it is also possible to obtain the cycle corresponding with each speed, such as: in 309 example,
Pir is divided into 3 speed, pir=108223*16kpbs, minimum message length l is 8bit for 1byte, tmax=500 μ s, then
In cycle tmaxMiddle added tokenmax=pir*tmax=108223*16kpbs*500 μ s=108223byte, period 1 t1
For 15.625 μ s, calculated according to 304-307, obtained first rate v1=108192*16kpbs, then can obtain token
Difference=tokenmax- v1*t1*[tmax/t1]=31byte, the second speed v2=30*16kpbs, third speed v3=1*16kpbs, will
Token difference 31byte is divided into 30byte and 1byte it is considered to by 30byte in tmaxIn be divided into 2 interpolations, by 1byte in tmaxIn
1 interpolation, then second round t2=30byte/2/(30*16kpbs)=250 μ s, period 3 t3=1byte/1/(1*16kpbs)=
500μs.
313, corresponding with the period 1 the first token interpolation quantity is obtained according to described first rate, according to described the
Two speed obtain second token corresponding with second round and add quantity.
Wherein, if pir is divided into more speed, such as third speed, fourth rate, then obtain also according to each speed
Obtain each cycle corresponding token and add quantity, such as threeth token corresponding with the period 3 is obtained according to third speed and add number
Amount, such as obtains fourth token corresponding with the period 4 according to fourth rate and adds quantity.
Further, it is also possible to arrange period 1 and second round in advance in the case of pir determination, or
To after first rate and the second speed it is also possible to historically the mapping table in speed and cycle to search with first rate,
Second speed corresponding period 1 and second round, thus obtain suitable period 1 and second round.
Further, the unit that network node can be safeguarded according to the demand planning token of qos, because token=pir*t,
T is the interpolation cycle of token, and token is that interpolation cycle corresponding token adds quantity, and the token that each cycle adds can be with table
It is shown as the form of n ' * m ', n ' is positive integer, the unit that m ' safeguards for token, and m ' can be less than bit, such as 0.25bit, m ' also may be used
With, such as 16byte bigger than byte.Such as: the token that each cycle adds is 3376byte, according to the demand planning token of qos
Safeguard unit m '=16byte, then token can write 211*16byte.Also such as: the token that each cycle adds is
4byte, unit m safeguarded according to the demand planning token of qos '=0.25bit, then token can write 128*0.25bit.
314, add quantity according to described period 1 and described first token and add token, according to described second round and
Described second token adds quantity and adds token.
Wherein, if pir is divided into more speed, such as third speed, fourth rate, appoint so according to divide each
Speed each cycle corresponding and each cycle corresponding token add quantity and add token, such as according to period 3 and the 3rd order
Board adds quantity and adds token, adds quantity according to period 4 and the 4th token and adds token.
The method that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir) the minimum refresh cycle that, and according to the hardware performance of network node can support sets the period 1, and pir is split into
First rate and the second speed, and according to tokenmaxObtain second round with the second speed, obtain and the further according to first rate
One cycle, corresponding first token added quantity, obtains second token corresponding with second round according to the second speed and adds number
Amount, adds quantity according to period 1 and the first token and adds token, adds quantity according to second round and the second token and adds
Token.Compared with from adding the prior art of token compared with large period, the scheme that the present invention provides, pir is divided in order to multiple
Different speed, and add token respectively according to each speed marking off so that adding according to the speed single that each marks off
Plus token be less than the token adding according to pir single, it is to avoid because needing of excessively being led to of the token that single adds passes
Defeated data burst is excessive, thus reducing peak flow, therefore improves qos；Simultaneously as the pir=1*pir of minimum
Precision, in cycle tmaxThe token of middle interpolationmaxBe not in fractional part, decrease depositing shared by the token that storage is added
The storage resource that storage capacity takies.
The embodiment of the present invention additionally provides the device 400 that a kind of token is safeguarded, as shown in Figure 5, comprising:
First acquisition module 401, for obtaining the peak information rate (pir) of network node.
Split module 402, for described pir is split into first rate and the second speed.
Wherein, the summation of first rate and the second speed is equal to pir.
Second acquisition module 403, adds for obtaining the first token corresponding with the period 1 according to described first rate
Addend amount, obtains second token corresponding with second round according to described second speed and adds quantity.
Add module 404, adds token for adding quantity according to described period 1 and described first token, according to institute
State second round and described second token adds quantity and adds token.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir), and by pir split into first rate and the second speed, obtain corresponding with the period 1 the further according to first rate
One token adds quantity, obtains second token corresponding with second round according to the second speed and adds quantity, according to the period 1
Add quantity with the first token and add token, add quantity according to second round and the second token and add token.With from larger
The prior art that cycle adds token is compared, the scheme that the present invention provides, and pir is divided for multiple different speed, and presses
Add token respectively according to each speed marking off so that token that the speed single that marks off according to each adds be less than according to
The token that pir single adds, it is to avoid because the data burst that needs that the token that single adds excessively is led to transmit is excessive,
Thus reducing peak flow, therefore improve qos.
Optionally, as shown in fig. 6, device 400 can also include:
3rd acquisition module 405, the minimum refresh cycle that the hardware performance for obtaining described network node can be supported.
Period 1 acquisition module 406, is institute for being more than or equal to the setting value of described minimum refresh cycle by one
State the period 1.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir), pir is split into first rate and the second speed, can also be supported according to the hardware performance of network node
The little refresh cycle chooses the period 1, obtains first token corresponding with the period 1 further according to first rate and adds quantity,
Second token corresponding with second round is obtained according to the second speed and adds quantity, add number according to period 1 and the first token
Amount adds token, adds quantity according to second round and the second token and adds token.Add showing of token with from compared with large period
Technology is had to compare, the scheme that the present invention provides, pir is divided for multiple different speed, and according to each speed marking off
Rate adds token respectively so that being less than, according to the token that the speed single that each marks off adds, the order added according to pir single
Board, it is to avoid because the data burst that needs that the token that single adds excessively is led to transmit is excessive, thus reduce flow
Peak value, therefore improves qos.
Optionally, may include that as shown in fig. 7, splitting module 402
First acquisition unit 4021, for obtaining l and k, and obtains n, n=l/(k*t1).
Wherein, l represents minimum message length, and k represents the precision of pir, t1Represent the period 1.
Second acquisition unit 4022, for obtaining w using described n.
Wherein, n=2w, or 2wIt is a smallest positive integral being more than n.
3rd acquiring unit 4023, for determining 0 to the m bit of described pir, and obtains w position to m position
Binary number.
First rate acquiring unit 4024, obtains described first rate for the binary number according to w position to m position.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir) the minimum refresh cycle that, and according to the hardware performance of network node can support chooses the period 1, and pir is split into
First rate and the second speed, obtain first token corresponding with the period 1 further according to first rate and add quantity, according to
Second speed obtains second token corresponding with second round and adds quantity, adds quantity according to period 1 and the first token and adds
Plus token, add quantity according to second round and the second token and add token.With the existing skill selected compared with large period interpolation token
Art is compared, the scheme that the present invention provides, and pir is divided for multiple different speed, and divides according to each speed marking off
Not Tian Jia token so that the token that the speed single that marks off according to each adds is less than the token adding according to pir single, keep away
The data burst having exempted to transmit by needs that excessively led to of token that single adds is excessive, thus reducing peak flow,
Therefore improve qos.
Optionally, can also include as shown in figure 8, splitting module 402:
4th acquiring unit 4025, in 0 to the m bit of described pir, obtaining w ' position to w-1 position
Binary number, wherein, 0≤w ' ＜ w.
Second speed acquiring unit 4026, obtains described second speed for the binary number according to w ' position to w-1 position
Rate.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir) the minimum refresh cycle that, and according to the hardware performance of network node can support chooses the period 1, and pir is split into
First rate and the second speed, obtain first token corresponding with the period 1 further according to first rate and add quantity, according to
Second speed obtains second token corresponding with second round and adds quantity, adds quantity according to period 1 and the first token and adds
Plus token, add quantity according to second round and the second token and add token.With the existing skill selected compared with large period interpolation token
Art is compared, the scheme that the present invention provides, and pir is divided for multiple different speed, and divides according to each speed marking off
Not Tian Jia token so that the token that the speed single that marks off according to each adds is less than the token adding according to pir single, keep away
The data burst having exempted to transmit by needs that excessively led to of token that single adds is excessive, thus reducing peak flow,
Therefore improve qos.
Optionally, as shown in figure 9, device 400 can also include:
4th acquisition module 407, for obtaining tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent described pir pair
The cycle answered, tmax≥l/k.
5th acquisition module 408, for obtaining token difference, described token difference=tokenmax- v1*t1*[tmax/t1],
Wherein v1Represent described first rate.
Second round acquisition module 409, for obtaining described second week using described token difference and described second speed
Phase.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir) the minimum refresh cycle that, and according to the hardware performance of network node can support sets the period 1, and pir is split into
First rate and the second speed, and according to tokenmaxObtain second round with the second speed, obtain and the further according to first rate
Corresponding first token of one cycle phase adds quantity, obtains second token corresponding with second round according to the second speed and adds number
Amount, adds quantity according to period 1 and the first token and adds token, adds quantity according to second round and the second token and adds
Token.Compared with from adding the prior art of token compared with large period, the scheme that the present invention provides, pir is divided in order to multiple
Different speed, and add token respectively according to each speed marking off so that adding according to the speed single that each marks off
Plus token be less than the token adding according to pir single, it is to avoid because needing of excessively being led to of the token that single adds passes
Defeated data burst is excessive, thus reducing peak flow, therefore improves qos；Simultaneously as the pir=1*pir of minimum
Precision, in cycle tmaxThe token of middle interpolationmaxBe not in fractional part, decrease depositing shared by the token that storage is added
The storage resource that storage capacity takies.
The embodiment of the present invention additionally provides the device 500 that a kind of token is safeguarded, as shown in Figure 10, comprising: at least one
Reason device 501, such as cpu, at least one communication bus 502, at least one network interface 503 and/or at least one user interface
504, memorizer 505.Communication bus 502 is used for realizing the connection communication between these assemblies.Network interface 503 can be used for passing
Transmission of data.User interface 504 can include the equipment such as display, keyboard, mouse.Memorizer 505 may comprise the storage of high speed ram
Device it is also possible to also include non-labile memorizer (non-volatile memory), for example, at least one disk memory.
Memorizer 505 can be used for storing the acquired peak information rate (pir) of network node, minimum message length and pir
Precision it is also possible to for storing the various parameters producing during processor processes data, can be also used for storing pir and splits into
Multiple speed and each cycle corresponding with each speed, and, can be also used for storing each cycle corresponding token
Add quantity.
Specifically, processor 501, for obtaining the peak information rate (pir) of network node；And, for will be described
Pir splits into first rate and the second speed, and wherein, the summation of described first rate and described second speed is equal to described pir；
And, add quantity for the first token corresponding with the period 1 is obtained according to described first rate, according to described second
Speed obtains second token corresponding with second round and adds quantity；And, for according to described period 1 and described first
Token adds quantity and adds token, adds quantity according to described second round and described second token and adds token.
Optionally, processor 501 is additionally operable to obtain the minimum refreshing week that the hardware performance of described network node can be supported
Phase；And, the setting value for being more than or equal to the described minimum refresh cycle by is the described period 1.
Further, processor 501 is additionally operable to obtain l and k, and obtains n, n=l/(k*t1), wherein, l represents tabloid
Civilian length, k represents the precision of pir, t1Represent the described period 1；And, for obtaining w, wherein n=2 using described nw, or
2wIt is a smallest positive integral being more than n；And, for determining 0 to the m bit of described pir, and obtain w position to m
The binary number of position；And, obtain described first rate for the binary number according to w position to m position.
Further, processor 501 is additionally operable in 0 to the m bit of described pir, acquisition w ' position to w-
The binary number of 1, wherein, 0≤w ' ＜ w；And, for obtaining described the according to the binary number of w ' position to w-1 position
Two speed.
Optionally, processor 501 is additionally operable to obtain tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent described pir
The corresponding cycle, tmax≥l/k；And, for obtaining token difference, described token difference=tokenmax- v1*t1*[tmax/
t1], wherein v1Represent described first rate；And, for obtaining described second using described token difference and described second speed
Cycle.
The device that a kind of token provided in an embodiment of the present invention is safeguarded, can obtain the peak information rate of network node
(pir) the minimum refresh cycle that, and according to the hardware performance of network node can support sets the period 1, and pir is split into
First rate and the second speed, and according to tokenmaxObtain second round with the second speed, obtain and the further according to first rate
Corresponding first token of one cycle phase adds quantity, obtains second token corresponding with second round according to the second speed and adds number
Amount, adds quantity according to period 1 and the first token and adds token, adds quantity according to second round and the second token and adds
Token.Compared with from adding the prior art of token compared with large period, the scheme that the present invention provides, pir is divided in order to multiple
Different speed, and add token respectively according to each speed marking off so that adding according to the speed single that each marks off
Plus token be less than the token adding according to pir single, it is to avoid because needing of excessively being led to of the token that single adds passes
Defeated data burst is excessive, thus reducing peak flow, therefore improves qos；Simultaneously as the pir=1*pir of minimum
Precision, in cycle tmaxThe token of middle interpolationmaxBe not in fractional part, decrease depositing shared by the token that storage is added
The storage resource that storage capacity takies.
Each embodiment in this specification is all described by the way of going forward one by one, identical similar portion between each embodiment
Divide mutually referring to what each embodiment stressed is the difference with other embodiment.Real especially for equipment
For applying example, because it is substantially similar to embodiment of the method, so describing fairly simple, referring to embodiment of the method in place of correlation
Part illustrate.
One of ordinary skill in the art will appreciate that realizing all or part of flow process in above-described embodiment method, it is permissible
Instruct related hardware to complete by computer program, described program can be stored in a computer read/write memory medium
In, this program is upon execution, it may include as the flow process of the embodiment of above-mentioned each method.Wherein, described storage medium can be magnetic
Dish, CD, read-only memory (read-only memory, rom) or random access memory (random access
Memory, ram) etc..
The above, the only specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, and any
Those familiar with the art the invention discloses technical scope in, the change or replacement that can readily occur in, all answer
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be defined by scope of the claims.
Claims (10)
1. a kind of method that token is safeguarded is it is characterised in that include:
Obtain the peak information rate pir of network node；
Described pir is split into first rate and the second speed, wherein, summation of described first rate and described second speed etc.
In described pir；
First token corresponding with the period 1 is obtained according to described first rate and adds quantity, obtained according to described second speed
Obtain second token corresponding with second round to add quantity, described period 1 and described second round, described first order
Board adds quantity and described second token adds quantity and is integer；
Add quantity according to described period 1 and described first token and add token, according to described second round and described second
Token adds quantity and adds token.
2. method according to claim 1 is it is characterised in that also include:
Obtain the minimum refresh cycle that the hardware performance of described network node can be supported；
One setting value being more than or equal to the described minimum refresh cycle it is the described period 1.
3. method according to claim 1 is it is characterised in that described split into first rate and the second speed by described pir
Rate includes:
Obtain l and k, and obtain n, n=l/ (k*t1), wherein, l represents minimum message length, and k represents the precision of pir, t1Represent
The described period 1；
Obtain w, wherein n=2 using described nw, or 2wIt is a smallest positive integral being more than n；
Determine 0 to the m bit of described pir, and obtain the binary number of w position to m position；
Binary number according to w position to m position obtains described first rate.
4. method according to claim 3 is it is characterised in that described split into first rate and the second speed by described pir
Rate also includes:
In 0 to the m bit of described pir, obtain the binary number of w ' position to w-1 position, wherein, 0≤w ' ＜ w；
Binary number according to w ' position to w-1 position obtains described second speed.
5. the method according to claim 3 or 4 is it is characterised in that also include:
Obtain tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent described pir corresponding cycle, tmax≥l/k；
Obtain token difference, described token difference=tokenmax- v1*t1*[tmax/t1], wherein v1Represent described first rate；
Obtain described second round using described token difference and described second speed.
6. the device that a kind of token is safeguarded is it is characterised in that include:
First acquisition module, for obtaining the peak information rate pir of network node；
Split module, for described pir is split into first rate and the second speed, wherein, described first rate and described the
The summation of two speed is equal to described pir；
Second acquisition module, adds quantity for obtaining the first token corresponding with the period 1 according to described first rate,
Second token corresponding with second round is obtained according to described second speed and adds quantity, described period 1 and described second week
Phase, described first token adds quantity and described second token adds quantity and is integer；
Add module, adds token for adding quantity according to described period 1 and described first token, according to described second
Cycle and described second token add quantity and add token.
7. device according to claim 6 is it is characterised in that also include:
3rd acquisition module, the minimum refresh cycle that the hardware performance for obtaining described network node can be supported；
Period 1 acquisition module, the setting value for being more than or equal to the described minimum refresh cycle by is described first week
Phase.
8. device according to claim 6 is it is characterised in that described fractionation module includes:
First acquisition unit, for obtaining l and k, and obtains n, n=l/ (k*t1), wherein, l represents minimum message length, and k represents
The precision of pir, t1Represent the described period 1；
Second acquisition unit, for obtaining w, wherein n=2 using described nw, or 2wIt is a smallest positive integral being more than n；
3rd acquiring unit, for determining 0 to the m bit of described pir, and obtains the binary system of w position to m position
Number；
First rate acquiring unit, obtains described first rate for the binary number according to w position to m position.
9. device according to claim 8 is it is characterised in that described fractionation module also includes:
4th acquiring unit, for, in 0 to the m bit of described pir, obtaining the binary system of w ' position to w-1 position
Number, wherein, 0≤w ' ＜ w；
Second speed acquiring unit, obtains described second speed for the binary number according to w ' position to w-1 position.
10. device according to claim 8 or claim 9 is it is characterised in that also include:
4th acquisition module, for obtaining tokenmax, tokenmax=pir*tmax, wherein tmaxRepresent described pir corresponding week
Phase, tmax≥l/k；
5th acquisition module, for obtaining token difference, described token difference=tokenmax- v1*t1*[tmax/t1], wherein v1
Represent described first rate；
Second round acquisition module, for obtaining described second round using described token difference and described second speed.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310514262.5A CN103560974B (en) | 2013-10-25 | 2013-10-25 | Method and device for maintaining tokens |
| US14/523,447 US9641440B2 (en) | 2013-10-25 | 2014-10-24 | Method and apparatus for maintaining token |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310514262.5A CN103560974B (en) | 2013-10-25 | 2013-10-25 | Method and device for maintaining tokens |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103560974A CN103560974A (en) | 2014-02-05 |
| CN103560974B true CN103560974B (en) | 2017-01-25 |
Family
ID=50015126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310514262.5A Active CN103560974B (en) | 2013-10-25 | 2013-10-25 | Method and device for maintaining tokens |
Country Status (2)
| Country | Link |
|---|---|
| US (1) | US9641440B2 (en) |
| CN (1) | CN103560974B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9654483B1 (en) * | 2014-12-23 | 2017-05-16 | Amazon Technologies, Inc. | Network communication rate limiter |
| CN105740076B (en) * | 2016-01-30 | 2019-06-21 | 华为技术有限公司 | A kind of load-balancing method and device |
| CN107948090B (en) * | 2017-11-23 | 2021-04-30 | 郑州云海信息技术有限公司 | Bandwidth control method and device for Ceph storage cluster volume |
| CN112242959B (en) * | 2019-07-16 | 2022-10-14 | 中国移动通信集团浙江有限公司 | Micro-service current-limiting control method, device, equipment and computer storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1949748A (en) * | 2006-10-09 | 2007-04-18 | 华为数字技术有限公司 | Method and apparatus for multiuser QOS dispatching |
| CN101778043A (en) * | 2010-01-19 | 2010-07-14 | 华为技术有限公司 | Method and device for dividing filling rate interval based on token bucket algorithm |
| CN102164083A (en) * | 2011-04-18 | 2011-08-24 | 中兴通讯股份有限公司 | Method and device for refreshing token bucket |
| CN103248580A (en) * | 2013-04-08 | 2013-08-14 | 汉柏科技有限公司 | High-precision token bucket speed limit method based on major-minor buckets |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9509921D0 (en) * | 1995-05-17 | 1995-07-12 | Roke Manor Research | Improvements in or relating to mobile radio systems |
| US5745490A (en) * | 1996-06-19 | 1998-04-28 | Pmc-Sierra, Inc. | Variable bit rate scheduler |
| US6801500B1 (en) * | 2000-05-18 | 2004-10-05 | Cisco Technology, Inc. | Method and apparatus for providing reserved rates to multiple flows on a network interface |
| EP1652391A2 (en) * | 2003-07-24 | 2006-05-03 | Koninklijke Philips Electronics N.V. | Admission control to wireless network based on guaranteed transmission rate |
| US20050078602A1 (en) * | 2003-10-10 | 2005-04-14 | Nortel Networks Limited | Method and apparatus for allocating bandwidth at a network element |
| US7602721B1 (en) * | 2003-12-22 | 2009-10-13 | Extreme Networks, Inc. | Methods and systems for fine grain bandwidth allocation in a switched network element |
| JP3961000B2 (en) * | 2005-05-26 | 2007-08-15 | 株式会社日立コミュニケーションテクノロジー | Packet transfer apparatus and network system |
| US7830796B2 (en) * | 2006-10-18 | 2010-11-09 | Ericsson Ab | Method and apparatus for traffic shaping |
| EP2211502A1 (en) * | 2009-01-22 | 2010-07-28 | IBBT vzw | Management system and method for configuring a network for transferring media data |
| US8289852B2 (en) * | 2009-02-18 | 2012-10-16 | Clearwire Ip Holdings Llc | Setting token bucket parameters for scheduling of air-interface resources |
| US8155003B2 (en) * | 2009-10-23 | 2012-04-10 | Cisco Technology, Inc. | Aggregate policing applying max-min fairness for each data source based on probabilistic filtering |
| US9219654B2 (en) * | 2010-06-25 | 2015-12-22 | Cox Communications, Inc. | Preloading token buckets for dynamically implementing speed increases |
| US9319325B2 (en) * | 2010-08-24 | 2016-04-19 | Intel Corporation | Adaptive method and system of regulation of yellow traffic in a network |
| EP2667662A4 (en) * | 2011-01-19 | 2016-05-18 | Fujitsu Ltd | Transmission device, transmission method and transmission system |
| EP2723021A1 (en) * | 2012-10-18 | 2014-04-23 | Telefonaktiebolaget L M Ericsson AB (Publ) | A method and an apparatus for determining the presence of a rate limiting mechanism in a network |
-
2013
- 2013-10-25 CN CN201310514262.5A patent/CN103560974B/en active Active
-
2014
- 2014-10-24 US US14/523,447 patent/US9641440B2/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1949748A (en) * | 2006-10-09 | 2007-04-18 | 华为数字技术有限公司 | Method and apparatus for multiuser QOS dispatching |
| CN101778043A (en) * | 2010-01-19 | 2010-07-14 | 华为技术有限公司 | Method and device for dividing filling rate interval based on token bucket algorithm |
| CN102164083A (en) * | 2011-04-18 | 2011-08-24 | 中兴通讯股份有限公司 | Method and device for refreshing token bucket |
| CN103248580A (en) * | 2013-04-08 | 2013-08-14 | 汉柏科技有限公司 | High-precision token bucket speed limit method based on major-minor buckets |
Non-Patent Citations (1)
| Title |
|---|
| 《QoS技术中令牌桶算法实现方式比较》;李晓利 等;《中兴通讯技术》;20070610;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| US20150117214A1 (en) | 2015-04-30 |
| CN103560974A (en) | 2014-02-05 |
| US9641440B2 (en) | 2017-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10841241B2 (en) | Intelligent placement within a data center | |
| CN103560974B (en) | Method and device for maintaining tokens | |
| US20150120923A1 (en) | Optimization Of Resource Utilization In A Collection of Devices | |
| CN108279974B (en) | Cloud resource allocation method and device | |
| US20150133214A1 (en) | Video encoding based on areas of interest | |
| CN104901989A (en) | Field service providing system and method | |
| CN102938790A (en) | Resource allocation method of cloud computing system | |
| CN103763343A (en) | Method and device for processing service access | |
| CN106130960B (en) | Judgement system, load dispatching method and the device of steal-number behavior | |
| CN104683408A (en) | Method and system for OpenStack cloud computing management platform to build virtual machine instance | |
| CN103746934A (en) | CDN bandwidth balancing method, CDN control center and system | |
| CN105183585B (en) | Data backup method and device | |
| TWI697223B (en) | Data processing method | |
| CN106303112B (en) | A kind of method for equalizing traffic volume and device | |
| CN107920108A (en) | A kind of method for pushing of media resource, client and server | |
| CN105808342A (en) | Method used for allocating client requests and judgment apparatus and system | |
| CN106375471A (en) | Edge node determination method and device | |
| WO2016197458A1 (en) | Traffic control method and apparatus | |
| CN104793982A (en) | Method and device for establishing virtual machine | |
| CN104202305A (en) | Transcoding processing method and device, server | |
| CN111221700B (en) | Cluster node state monitoring method, device, equipment and readable storage medium | |
| CN114138468B (en) | Self-adaptive distribution method and device for packaging task amount and storage medium | |
| CN108153494B (en) | A kind of I/O request processing method and processing device | |
| CN104168174A (en) | Method and apparatus for information transmission | |
| CN109495531A (en) | Request processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |